(11.2) (11.1) Simultaneously, the dry bulk density is obtained through (Equation 3.5) (11.3) where pb = the dry bulk density (kg/m3)

Transcription

1 11 Water in the Unsaturated Zone P Kabat' and J Beekma2 111 Introduction In the soil below the watertable, all the pores are generally filled with water and this region is called the saturated zone When, in a waterlogged soil, the watertable is lowered by drainage, the upper part of the soil will become unsaturated, which means that its pores contain both water and air Water in the unsaturated zone generally originates from infiltrated precipitation and from the capillary rise of groundwater The process of water movement in the unsaturated part of the soil profile plays a central role in studies of irrigation, drainage, evaporation from the soil, water uptake by roots, and the transport of salts and fertilizers The unsaturated zone is of fundamental importance for plant growth Soil-water conditions in the upper part of the soil profile have a distinct influence on the accessibility, trafficability, and workability of fields A knowledge of the physical processes in the unsaturated zone is essential for a proper estimate of drainage criteria and for evaluating the sustainability of drainage systems This chapter introduces some basic soil physics concerning the movement of water in unsaturated soil, and gives some examples of their use in drainage studies Several methods of measuring soil-water status and soil hydraulic parameters are dealt within Sections 112 and 113 Basic relations and parameters governing water flow in the unsaturated zone are explained in Sections 1 14 and 1 15 This is followed by a discussion of the extraction of water through plant roots (Section 116) Section 1 17 treats the preferential flow of water through unsaturated soil The steady-state approach is illustrated with the help of a computer program; the unsteady-state flow is highlighted with a numerical simulation model (Section 118) The model combines unsaturated-zone dynamics with the characteristics of a drainage system This enables us to evaluate the effects of a drainage system on soil-water conditions for crop production and on solute transport through the soil 112 Measuring Soil-Water Content The main constituents of soil are solid particles, water, and air They can be expressed as a fraction or as a percentage Basic formulas for soil water content on a volume basis and on a mass basis were given in Chapter 3 (Equations 31 to 39) In practice, one often expresses soil-water content over a depth of soil directly in mm of water Thus, 8 = 010 means that 10 mm of water is stored in a 100 mm soil column (010 x 100 = 10) Soil-water content can be measured either with destructive methods or with non-destructive methods An advantage of non-destructive measurements is ' The Winand Staring Centre for Integrated Land, Soil and Water Research International Institute for Land Reclamation and Improvement 383

2 that repetitive measurements can be taken at the same location This advantage becomes most pronounced when we combine it with automatic data recording The gravimetric method, which leads to a soil-water content on the basis of weight or volume, is the most widely used destructive technique Non-destructive techniques that have proved to be applicable under field conditions are: - Neutron scattering; - Gamma-ray attenuation; - Capacitance method; - Time-domain reflectrometry Gravimetric Method A soil sample is weighed, then dried in an oven at 105 C, and weighed again The difference in weight is a measure of the initial water content Samples can be taken on a mass or on a volume basis In the first case, we take a disturbed quantity of soil, put it in a plastic bag, and transport it to the laboratory, where it is weighed, dried, and re-weighed after drying We calculate the mass fraction of water with where w = fraction of water on mass base (kgkg- ) m, = mass of water in the soil sample (kg) m, = mass of solids in the soil sample (oven dry soil) (kg) (111) To get the soil-water content on a volume basis, we need samples of known volume We normally use stainless steel cylinders (usually 100 cm ), which are pushed horizontally or vertically into profile horizons We subsequently retrieve and trim the filled cylinder, and put end caps on Soil horizons are exposed in a soil pit If no pit can be dug, we can use a special type of auger in which the same type of cylinder is fixed The volume fraction of water can be calculated as where 8 = volumetric soil-water content (m3m- ) V = volume of cylinder (m ) pw = density of water (kg/m3); often taken as 1000 kg/m3 V, = volume of water (m ) Simultaneously, the dry bulk density is obtained through (Equation 35) (112) (113) where pb = the dry bulk density (kg/m3) 384

3 We can convert the soil-water content on mass base (w) to a volumetric soil-water content (0) (114) The gravimetric method is still the most widely used technique to determine the soilwater content and is often taken as a standard for the calibration of other methods A disadvantage is that it is laborious, because samples in duplicate or in triplicate are required to compensate for errors and variability Moreover, volumetric samples need to be taken carefully The samples cannot usually be weighed in the field, and special care must be taken to prevent them from drying out before they are weighed in the laboratory Neutron-Scattering The neutron-scattering method is based on fast-moving neutrons emitted by a radioactive source, usually 241Am, which collide with nuclei in the soil and lose energy A detector counts part of the slowed-down reflected (thermal) neutrons Because hydrogen slows down neutrons much more than other soil constituents, and since hydrogen is mainly present in water, the neutron count is strongly related to the water content We use an empirical linear relationship between the ratio of the count to a standard count of the instrument, which is called the count ratio, and the soil-water content The standard count is taken under standard conditions, preferably in a pure water body The empirical relationship reads where 0=a+bR (1 15) R = the count ratio (-) a and b = soil specific constants (-) Because, apart from hydrogen, the count ratio is also influenced by the bulk density of the soil and by various chemical components, a soil specific calibration is required Constant a in Equation 115 increases with bulk density; constant b is influenced by soil chemical composition (Gardner 1986) The calibration can be done by regression of the soil-water content of samples taken around the measurement site, on the count ratio Calibration can also be done in a drum in the laboratory, but this is more cumbersome, since one needs to create soil conditions comparable to those in the field For field measurements, portable equipment has been developed The most frequently used equipment consists of a probe unit and a scaler (Figure 111) The probe, containing a neutron source, is lowered into a tube, called an access tube, in the soil down to the required depth A proportion of the reflected slow neutrons is absorbed in a boron-trifluoride gas-filled tube (counter) Ionization of the gas results in discharge pulses, which are amplified and measured with the scaler The action radius of the instrument is spherical and its size varies with soil wetness; the drier the soil, the larger the action radius (between approximately 15 cm in wet soil to 50 cm in dry soil) For a comparison of measurements from different locations, the size, shape, and material of the access tubes must be identical Aluminium is a frequently used material 385

4 scaler and counter recorder accesstube::: : $IOW neutrons 1 : : fast neutrons 1 : iadio-active source ' ''_''I t Figure 111 Neutron probe to measure soil-water content because it offers practically no resistance to slow neutrons; polyvinyl chloride (PVC), polythene, brass, and stainless steel show a lower neutron transmission For more details, see Gardner (1986) The neutron-scattering technique has been widely used under field conditions Advantages of the method are: - Soil-water content can be measured rapidly and repeatedly in the same place; - Average soil-water content of the sphere of influence can be measured with depth; - Temporal soil-water content changes can easily be followed; - Relation between count ratio and soil-water content is linear Disadvantages are: - Counts have a high variability; measurementsare not completely repeatable; - Poor depth resolution; - Measurements are interfered with by many soil constituents; - The use of a radioactive source can pose health risks if no appropriate care is taken and create disposal problems after use; - Measurements near the soil surface are impossible Gamma-Ray Attenuation With the gamma ray method, we can measure the soil's wet bulk density (see Chapter 3) If the dry bulk density does not change over the period considered, changes in wet bulk density are only due to changes in soil-water content If a beam of gamma rays emitted by a Cesium'37 source is transmitted through the soil, they are attenuated (reduced in intensity), the degree of attenuation increasing with wet bulk density (Bertuzzi et al 1987) The field method (Figure 112) requires two access tubes, one for the source and one for the detector These access tubes must be injected precisely parallel and vertical, because the gamma method is highly- susceptible to deviations in distance Sometimes two gamma-ray sources with different energies are used With such a dual-source 386

5 injoutput connection 1 guiding system locking system Annulus container \ /A\W/A\Y I P alignement jig I- source tube? aluminium : 1' detecto! tub,e (16 mm OD) : access tubes A D : (40mmOD) ' (32mmOD) I ] : ::: ':: 'radioactive source' detector' - 1 : C"(5 "1!), ::': ' ' cr]stalnal(ti! : Figure 112 The gamma-ray probe (after Bertuzzi et al 1987) method (Gurr and Jakobsen 1978), dry bulk density and water content are obtained separately The method is especially suitable for swelling soils The calibration procedure depends on the type of instrument For details, see Gurr and Jakobsen (1978) and Gardner (1986) The method is less widely applied than the neutron-scattering method, and is mostly used to follow soil-water content in soil columns in the laboratory The advantage of the method is that the data on soil-water content can be obtained with good depth resolution Disadvantages are: - Field instrumentation is costly and difficult to use; - Extreme care must be taken to ensure that the radioactive source is not a health hazard Capacitance Method The capacitance method is based on measuring the capacitance of a capacitor, with the soil-water-air mixture as the dielectric medium The method has been described 387

6 by, among others, Dean et al (1987) Its application and accuracy under field conditions was investigated by Halbertsma et al (1987) A probe with conductive plates or rods surrounded with soil constitutes the capacitor The relative permittivity (dielectric constant) of water is large compared with that of the soil matrix and air A change in the water content of the soil will cause a change in the relative permittivity, and consequently in the capacitance of the capacitor (probe) surrounded with soil The capacitor is usually part of a resonance circuit of an oscillator Changes in the soil-water content, and thus changes in the capacitor capacitance, will change the resonance frequency of the oscillator In this way, the water content is indicated by a frequency shift Since the relative permittivity of the soil matrix depends on its composition and its bulk density, calibration is needed for each separate soil The field instrument consists of a read-out unit and either a mobile probe to be able to measure in different access tubes or fixed probes (Hilhorst 1984) installed at different depths within the soil profile (Figure 1 13) The capacitance method has been used with good results in several studies Generally, the accuracy of determining the soil-water content was reported to be in the range of f 002 (m3m-') (Halbertsma et al 1987) This accuracy is limited by the calibration, rather than by the instrument or by the measurement technique itself The capacitive instrument can be combined with an automatic data recording system Such a system can collect soil-water data more or less continuously The advantages of the method are comparable to those of the neutron-scattering method Additional advantages are: - Good depth resolution; - Very fast response; O 2 4 6cm - Figure 1 I 3 An example of the installation of the capacitance probes in a soil profile and a schematic illustration of the capacitance probe (after Halbertsma et al 1987) A The probes are placed in two columns in between two rows of a crop at different depths ranging between 2 and 60 cm B The capacitance probe consists of (a) a holder, (b) three electrodes, (c) a cable, and (d) a connector 388

7 - Little diversion of measured frequency for repeated measurements; - Different portable versions are available for field use; - The instrument is inherently safe; - It can be combined with an automated data-recording system; - Surface soil-water content can be measured Disadvantages are: - Relationship between frequency shift and soil-water content is non-linear; - The method is sensitive to electrical conductivity of the soil; - The installation of access tube or probe has to be done with care; small cavities around the tube have a great influence on the measured frequency Time-Domain Reflectrometry A method that also uses the dielectrical properties of the soil is time-domain reflectrometry (TDR) The propagation time of a pulse travelling along a wave guide is measured This time depends on the dielectrical properties of the soil surrounding the wave guide, and hence on the water content of the soil The TDR method can be used for many soils without calibration, because the relationship between the apparent dielectric constant and volumetric water content is only weakly dependent on soil type, soil density, soil temperature, and salt content (Topp and Davis 1985) Topp et al (1980) reported a measured volumetric water content with an accuracy of k 002 (m3m- ) Time-domain reflectrometry has become popular in recent years, mainly because the method does not need elaborate calibration procedures Several portable, batterypowered TDR units are available at this moment Electrodes to be used as the actual measuring device are available in different configurations The full potential of this method is only realized when it is combined with an automatic data acquisition system (eg Heimovaraa and Bouten 1990) The advantages of TDR are comparable to those of the capacitance method Additional advantages are: - Highly accurate soil-water content measurements at desired depths; - Availability of electrodes with required ranges of influence; - No calibration required for different soil types Disadvantages are: - Expensive electrodes and data-recording systems, resulting in high costs if an extensive spatial coverage is desired; - Electrodes difficult to install in stony and heavily compacted soils 113 Basic Concepts of Soil-Water Dynamics To describe the condition of water in soil, mechanical and thermo-dynamic (or energy) concepts are used In the mechanical concept, only the mechanical forces moving water through the soil are considered It is based on the idea that, at a specific point, water in unsaturated soil is under a pressure deficit as compared with free water In the energy concept, other driving forces are considered in addition to mechanical forces These forces are caused by thermal, electrical, or solute-concentration gradients 389

8 1131 Mechanical Concept The mechanical concept can be illustrated by regarding the soil as a mixture of solids and pores in which the pores form capillary tubes If such a small capillary tube is inserted in water, the water will rise into the tube under the influence of capillary forces (Figure 1 14) The total upward force lifting the water column, Ff, is obtained by multiplying the vertical component of surface tension by the circumference of the capillary where Ff = ocosa x 2nr Ff = upward force (N) o = surface tension of water against air (o = 0073 kgs-* at 20 C) a contact angle of water with the tube (rad); (cos a N 1) r = equivalent radius of tube (m) (116) By its weight in the gravitational field, the water column of length h and mass m2hp exerts a downward force F1 that opposes capillary rise FJ = nr2hp x g where FJ= downward force (N) p = density of water (p = 1000 kg/m3) g = acceleration due to gravity (g = 981 m/s2) h = height of capillary rise (m) (1 17) Figure 1 14 Capillary rise of water 390

9 At equilibrium, the upward force Ff equals the downward force F1 and water movement stops In that case or ocosa x 2nr = nr2hp x g h= 20 cos a Pgr (118) Substituting the values of the various constants leads to the expression for the height of capillary rise h=- 015 r Thus the smaller the tube, the higher the height of capillary rise (119) 1132 Energy Concept Real soils do not consist of capillaries with a characteristic diameter Water movement in soil, apart from differences in tension, is also caused by thermal, electrical, or soluteconcentration gradients The forces governing soil-water flow can accordingly be described by the energy concept According to this principle, water moves from points with higher energy status to points with lower energy status The energy status of water is simply called 'water potential' The relationship between the mechanical- force concept and the energy-water-potential concept is best illustrated for a situation in which the distance between two points approximates zero The forces acting on a mass of water in any particular direction are then defined as (1 110) where Fs = total of forces (N) m = mass of water (kg) s = distance between points (m) $ = water potential on mass base (J/kg) The negative sign shows that the force works in the direction of decreasing water potential The water potential is an expression for the mechanical work required to transfer a unit quantity of water from a standard reference, where the potential is taken as zero, to the situation where the potential has the defined value Potentials are usually defined relative to water with a composition identical to the soil solution, at atmospheric pressure, a temperature of 293 K (20 C), and datum elevation zero Total water potential, $t, is the sum of several components (Feddes et al 1988) $t = $In + $ex + $en + $s + $, + (1111) 39 1

10 where Qt = total water potential $, = matrix potential, arising from local interactions between the soil matrix and water $ex = excess gas potential, arising from the external gas pressure $en = envelope or overburden potential, arising from swelling of the soil Jr, = osmotic potential, arising from the presence of solutes in the soil water = gravitational potential, arising from the gravitational force Jr, In soil physics, water potential can be expressed as energy on a mass basis (I)"), on a volume basis (v), or on a weight basis ($") As an example, let us take the gravitational potential, $g, with the watertable as reference level The definition of potential says that the mechanical work required to raise a mass of water (m = pv) from the watertable to a height z is equal to mgz or pvgz Thus the gravitational potential on mass basis ($gm), on volume basis ($i), or on weight basis ($,") will be $: = = gz (J/kg) PV (1112) (1113) (1114) We can do the same for other potentials The general relationship of potentials based on mass (I)"), on volume (v), and on weight ($") is $":qp:$"=g:pg:l (1115) This means that the values of $"' are a factor 98 higher than corresponding values of QW; values of I)" are a factor 9800 higher (for p water = lo3 kg/m3), for which reason we often use kpa as a unit of qp instead of Pa In hydrology, one prefers to use the potential on a weight basis, and potentials are referred to as 'heads' In the following, we shall restrict ourselves to water potentials based on weight In analogy to Equation , we can write h, = h, + hex + he, + h, + h, + (1116) with the potentials now called 'heads' and the subscripts having the same meaning as in Equation : - The matric head (h,) in unsaturated soil is negative, because work is needed to withdraw water against the soil-matric forces At the groundwater level, atmospheric pressure exists and therefore h, = O; - Changes in total water head in the soil may also be caused by changes in the pressure of the air adjacent to it In natural soils, however, such changes are fairly exceptional, so we can assume that he, = O; - A clay soil that takes up water and swells will exert an additional pressure, he,, on the total water head In soils with a rigid matrix (non-swelling soils), he, = O; 3 92

11 - In soil-water studies, we can very often neglect the influence of the osmotic head, h, This is justified as far as we measure the head values relative to groundwater with the same or nearly the same chemical composition as the soil water; thus h, N O Where considerable differences in solute concentration in the soil profile exist, it is obviously necessary to take h, into account; - The gravitational head, h,, is determined at each point by the elevation of that point relative to a certain reference level Equation 1114 shows that h, = z, with z positive above the reference level and negative below it The sum of the components h,, he,, and he, is usually referred to as soil water pressure head, h, which can be measured with a tensiometer h = h, + he, + he, If we assume that he, and he, are zero, as mentioned earlier, we can write (1117) h, = h, + h, + h, (1118) Taking h, = O, h, = z and denoting h, as H, we can also write where H=h,+z (1119) H = hydraulic head (m) z = elevation head or gravitational head (m) According to Equation 1110, differences in head determine the direction and the magnitude of soil-water flow When the soil water is in equilibrium, 4H/& = O, and there is no flow Such a situation is shown in Figure 115, where the watertable height above reference level in cm 100 soil surface O I tensiomete! 1, : ::: - --li 21 hl soil'''''''' ' 1;; ~~;:::: ::::: ::: ::: ::: Figure 115 Equilibrium (no-flow) conditions in a soil profile with a watertable depth of 1 O m 393

12 is at 100 m depth, and the reference level is taken at this depth The pressure head in the soil is measured with tensiometers (For details on the functioning of tensiometers, see Section 1133) Tensiometer 1 is installed at 50 cm depth, and Tensiometer 2 at 140 cm depth The pressure head at the watertable is, by definition, h = p/pg = O, because the water there is in equilibrium with atmospheric pressure Above the watertable, h < O; below it h > O ( hydrostatic pressure ) For Tensiometer 1, the pressure head is represented by the height of the open end of the water column, h, = -50 cm, and gravitational head by the height above reference level, z, = 50 cm Thus H, = h, + z, = = Ocm In the same way, for Tensiometer 2, we find h, = 40cmandq = -4Ocm,thusH, = = O Hence, everywhere in the soil column, H = O cm and equilibrium exists and no water flow takes place The distribution of the pressure head and the gravitational head in a profile under equilibrium conditions is shown in Figure Measuring Soil-Water Pressure Head Techniques to measure soil-water pressure head, h, or the matric head, h,, are usually restricted to a particular range of the head We can use the following techniques: 1) Tensiometry for relatively wet conditions (-800 cm < h < O cm); height above reference level in cm 1 O O soil surface,, O Figure 116 Distribution of the soil-water pressure heads with depths under equilibrium conditions 394

13 2) Electrical resistance blocks for the range of -10 O00 cm < h < -20 cm; 3) Soil psychrometry for dry conditions (h < cm); 4) Thermal conductivity techniques (-3000 cm < h < -100 cm); 5) Techniques based on dielectrical properties ( cm < h < -10 cm) For practical field use, Techniques 3), 4), and 5) are not yet fully operational The soil-psychrometry method (Bruckler and Gaudu 1984) is difficult to perform since we need to achieve a thermal equilibrium between the sensor and the surrounding soil Thermal-conductivity-based techniques (Phene et al 1987) and the dielectrical method (Hilhorst 1986) are promising, but are not yet operational In field practice, tensiometry and, to a lesser extent, electrical resistance blocks are mainly used Tensiometry A tensiometer consists of a ceramic porous cup positioned in the soil This cup is attached to a water-filled tube, which is connected to a measuring device As long as there is a pressure-head gradient between the water in the cup and the water in the soil, water will flow through the cup wall Under equilibrium conditions, the pressure head of the soil water is obtained from the water pressure inside the tensiometer As the porous cup of the tensiometer allows air to enter the system for h < -800 cm, direct measurements of the pressure head in the field are only possible from O to -800 cm The principle of tensiometry can be seen in Figure 1 15 The soil profile is in hydrological equilibrium here, which means that at any place in the profile the pressure head (h) is equal to the reversal of the gravitational head (see also Figure 1 16), ie h = -z At measurement position 1 (tensiometer - cup l), a suction (-hl) draws the water in the tensiometer to the position where this suction is fully counteracted by the gravitational head, zi Hence, h, + zi = O and the measured pressure head has a negative value equal to -zi The pressure head is always negative in the unsaturated zone, which makes water tensiometers as in Figure 115 impractical When the conditions are not in equilibrium and if, say, h were lower than -2, a pit would have to be dug to read pressure head h Commonly used tensiometers are illustrated in Figure 117 They are: - Vacuum gauge (Type A); - Mercury-water-filled tubes (Types B and C) For Type B, we see that h = d,- (p,/pw)d, With the densities of mercury, pm = kg/m3, and water, pw = 1000 kg/m3, it follows that h = d,- 136 d, For Type C,,we see that h = d, - (p,/p,)d, and d, = do + d,, so that h = do + d,(l - p,/pw) M d0-126 d,; - Electronic transducers (Type D); they convert changes in pressure into small electrical forces, which are first amplified and then measured with a voltmeter We often use absolute values of the pressure head, Ihl, which, in daily practice, are called tensions or suctions of the soil A tension and a suction thus always have a positive value The setting-up time, or response time, of a tensiometer, defined as the time needed to reach equilibrium after a change in hydraulic head, is determined by the hydraulic 395

14 Figure 1 I 7 Tensiometers conductivity of the soil, the properties of the porous cup, and, in particular, by the water capacity of the tensiometer system The water capacity is related to the amount of water that must be moved in order to create a head difference of 1 cm The setting-up time of tensiometers with a mercury manometer or Bourdon manometer ranges from about 15 minutes in permeable wet soil to several hours in less permeable, drier soils Rapid variations in pressure head cannot be followed by a tensiometer Shorter settingup times can be obtained with manometers of small capacity This requirement can be met with the use of electrical pressure transducers Good contact between the soil and the porous cup of a tensiometer is essential for the functioning of a tensiometer The best way to place a tensiometer in the soil is to bore a hole with the same diameter as the porous cup to the desired depth and then to push the cup into the bottom of thehole Usually, tensiometers are installed permanently at different depths They can be connected by a distribution system of tubes and stopcocks to one single transducer The tensiometers can then be measured one by one Tensiometers have also been successfully combined with an automatic data-acquisition system (eg Van den Elsen and Bakker 1992) Electrical Resistance Blocks The principle of measuring soil-water suction with an electrical resistance block placed in the soil is based on the change in electrical resistance of the block due to a change in water content of the block The blocks consist of two parallel electrodes, embedded in gypsum, nylon, fibreglass, or a combination of gypsum with nylon or fibreglass The electrical resistance is dependent on the water content of the unit, the pressure head of which is in equilibrium with the pressure head of the surrounding soil It can be measured by means of a Wheatstone bridge and should be calibrated against the pressure head measured in an alternative way 396

15 ~ 1134 Electrolytes in the soil solution will give reduced resistance readings With gypsum blocks, however, this lowering of the resistance is counteracted by the saturated solution of the calcium sulphate in the blocks Application is therefore possible in slightly saline soils Contact between resistance unit and soil is essential, which restricts its use to nonshrinking soils In some sandy soils, where the pressure head changes very little with considerable change in soil-water content, measurements are inaccurate Soil-Water Retention The previous sections showed that the pressure head of water in the unsaturated soil arises from local interactions between soil and water When the pressure head of the soil water changes, the water content of the soil will also change The graph representing the relationship between pressure head and water content is generally called the soil-water retention curve or the soil-moisture characteristic As was explained in Chapter 3, applying different pressure heads, step by step, and measuring the moisture content allows us to find a curve of pressure head, h, versus soil-water content, 8 The pressure heads vary from O cm (for saturation) to lo7 cm (for oven-dry conditions) In analogy with ph, pf is the logarithm of the tension or suction in cm of water Thus pf = log Ihl (1120) Figure 118 shows typical water retention curves of four standard soil types Saturation The intersection point of the curves with the horizontal axis (tension: 1 cm water, pressure head in cm O +sand+ -loam- O I volumetric soil water content 1 range of available water 1 peat -_I +-clay+ Figure 1 I 8 Soil-water retention curves for four different soil types, and their ranges ofplant-available water 397

16 pf = O) gives the water content of the soils under nearly saturated conditions, which means that this point almost indicates the fraction of total pore space or porosity, E (Chapter 3) Field Capacity The term field capacity corresponds to the conditions in a soil after two or three days of free drainage, following a period of thorough wetting by rainfall or irrigation The downward flow becomes negligible under these conditions For practical purposes, field capacity is often approximated by the soil-water content at a particular soil-water tension (eg at 100,200, or 330 cm) In literature, soil-water tensions at field capacity range from about 50 to 500 cm (pf = 07-27) In the following, we shall take h = -100 cm (pf 20) as the fieldcapacity point It is regarded as the upper limit of the amount of water available for plants The air content at field capacity, called aeration porosity, is important for the diffusion of oxygen to the crop roots Generally, if the aeration porosity amounts to 10 or 15 vol% or more, aeration is satisfactory for plant growth Wilting Point The wilting point or permanent wilting point is defined as the soil water condition at which the leaves undergo a permanent reduction in their water content (wilting) because of a deficient supply of soil water, a condition from which the leaves do not recover in an approximately saturated atmosphere overnight The permanent wilting point is not a constant, because it is influenced by the plant characteristics and meteorological conditions The variation in soil-water pressure head at wilting point reported in literature ranges from to -30 O00 cm (Cassel and Nielsen 1986) In the following, we shall take h = -1 5 O00 (pf 42) as the permanent wilting point For many soils, except for the more fine-textured ones, a change in soil-water content becomes negligible over the range cm to -30 O00 cm (Cassel and Nielsen 1986) Oven-Dry Point When soil is dried in an oven at 105 C for at least 12 hours, one assumes that no water is left in the soil This point corresponds roughly with pf 7 Available Water The amount of water held by a soil between field capacity (pf 20) and wilting point (pf 42) is defined as the amount of water available for plants Below the wilting point, water is too strongly bound to the soil particles Above field capacity, water either drains from the soil without being intercepted by roots, or too wet conditions cause aeration problems in the rootzone, which restricts water uptake The ease of water extraction by roots is not the same over the whole range of available water At increasing desiccation of the soil, the water uptake decreases progressively For optimum plant production, it is better not to allow the soil to dry out to the wilting point The admissible pressure head at which soil water begins to limit plant growth varies between 400 and cm (PF 26 to pf 3) For most soils, the drought limit is reached when a fraction of 040 to 060 of the total amount of water available in 398

17 Table 111 The average amount of available water in the rootzone Soil type Coarse sand Medium coarse sand Medium fine sand Fine sand Loamy medium coarse sand Loamy fine sand Sandy loam Loess loam Fine sandy loam Silt loam Loam Sandy clay loam Silty clay loam Clay loam Light clay Silty clay Basin clay Peat Total available the rootzone is used This fraction is often referred to as readily available soil water From Figure 118, it is obvious that the absolute amount of water available in the rootzone depends strongly on the soil type Table 111 presents the average amounts of available water for a number of soils as derived from data in literature Hysteresis We usually determine soil-water retention curves by removing water from an initially wet soil sample (desorption) If we add water to an initially dry sample (adsorption), the water content in the soil sample will be different at corresponding tensions This phenomenon is referred to as hysteresis Due to the hysteresis effect, the watercontent-tension relationship of a soil depends on its wetting or drying history Under field conditions, this relationship is not constant The effect of hysteresis on the soilwater retention curve is shown in Figure 119A The hysteresis effect may be attributed to: - The pores having a larger diameter than their openings This can be explained by Equation 119, which not only holds for capillary rise, but also for the soil-water tension, h, as related to the pore diameter During wetting, the large pore will only take up water when the tension is in equilibrium with, or lower than, the tension related to its large diameter During drying, the pore opening diameter determines the tension needed to withdraw the water from the pore This tension should be higher than the tension calculated with Equation 119 The effect of this is illustrated in Figure 119B; - Variations in packing due to a re-arrangement of soil particles by wetting or drying; 399

18 - volumetric soil water content Figure 119 Hysteresis A In a family of pf-curves for a certain soil B Pore geometry as the phenomenon causing hysteresis - Incomplete water uptake by soils that have undergone irreversible shrinking or drying (some clay and peat soils); - Entrapped air Methods for Determining Soil- Water Retention Soil hydraulic conductivity (Section 115) and soil-water retention are the most important characteristics in soil-water dynamics Theoretically, if one were able to reproduce exactly the measurements on the same soil sample, and if natural soils were not spatially heterogeneous, each soil type would be characterized by one unique set of functions for soil-water retention and soil hydraulic conductivity Various methods have been developed to determine these characteristics, either in the laboratory or in situ The methods can be divided into direct and indirect approaches (Kabat and Hack-ten Broeke 1989) The indirect approaches to estimate, both soil-water retention and hydraulic conductivity will be presented in Section 1152 Below, only the basic principles of the direct measurements of soil-water retention will be discussed In-Situ Determinations Section 1 12 presented a number of operational methods to measure the volumetric soil-water content, and Section 1133 described techniques to measure the soil-water pressure head If we combine both measurements for the same place and time (ie with equipment installed in the same soil profile), we obtain an in-situ relationship between measured pressure head and volumetric soil-water content 400

19 Figure 1 1 IO Measurement of the soil-water characteristic in the range of 150 < h < O cm Laboratory Methods To determine the water retention of an undisturbed soil sample, the soil water content is measured for equilibrium conditions under a succession of known tensions Ihl Porous-Medium Method A soil sample cannot be exposed directly to suction because air will then enter and prevent the removal of water from the sample A water-saturated porous material is therefore used as an intermediary The porous medium should meet the following requirements: - It must be possible to apply the required suction without reaching the air-bubbling pressure (air-entry value), the pressure at which bubbles of air start to leak through the medium; - The water permeability of the medium has to be as high as possible, which is contradictory to the first requirement This demands a homogeneous pore-size distribution, matching the applied pressure Tension Range O cm Undisturbed volumetric soil samples are placed upon a porous medium that is watersaturated (Figure 1110) A water column of a certain length is then used to exert the desired suction or tension on the soil sample, via the porous medium As the pore-size distribution of the soil influences its water-retaining properties, undisturbed soil samples have to be used This method is called the hanging watercolumn method Tension Range cm A slightly different procedure is used in this range, instead of a hanging water column, suction is created by a vacuum line connected to ceramic plates The same volumetric samples are placed on these plates and water is drained from the samples until equilibrium with the plates is reached This method is called the suction plate method 40 1

20 soil sample I pressure regulator I atmospheric 1 L porous compressor pressure ceramic plate Figure Measurement of the soil-water characteristic in the range of < h < -500 cm Tension Range O00 cm In the range of 500 to 15 O00 cm, instead of applying suctions, pressures are exerted on the soil sample, which is placed on a porous medium in a chamber (Figure ) For pressures up to 3000 cm, undisturbed samples are normally used; for higher pressures, disturbed soil samples can be used As porous medium, a ceramic plate or a cellophane membrane is used Under the membrane, a shallow water layer under atmospheric pressure (zero gauge pressure) is present According to Equation 1 117, when he, is assumed to be zero, h = h, + he, Around the sample, the external imposed gas pressure is, say, 12 bar (ie equivalent to a head he, = 12 O00 cm) Water is discharged from the sample through the membrane into the water layer until equilibrium is reached Then the pressure inside the soil sample is atmospheric, h = O Hence, it follows that, or O = h, h, = -12 O00 cm With this method, h,- 0 relationships can be determined over a large range of tensions The method is referred to as the pressure pan method for the lower range, when ceramic plates are used, and as the pressure membrane method for higher pressures (Klute and Dinauer 1986) In the very dry range, for h < -30 O00 cm (pf > 49, the vapour pressure method can be applied For details, see Campbell and Gee (1986) 1135 Drainable Porosity The storage coefficient, p, also called drainable pore space, is important for unsteady drainage equations and for the calculation of groundwater recharge The storage coefficient is a constant that represents the average change in the water content of the soil profile when the watertable level changes with a discrete step Its value depends on soil properties and the depth of the watertable To derive a practical mean value of a storage coefficient for an area, it should be calculated for the major 402

21 soil series and for several depths of the watertable If the water retention of the soil is known and if the pressure-head profile is known for two different watertable levels, the storage coefficient p can be calculated from the following equation 72 Zl where zi = watertable depth for Situation 1 (m) z2 = watertable depth for Situation 2 (m) 8,(z) '= soil-water content as a function of soil depth for Watertable 1 (-) 8,(z) = soil-water content as a function of soil depth for Watertable 2 (-) (1121) I Usually, the drainable pore space is calculated for equilibrium conditions between soil-water content and watertable depth The computer program CAPSEV (Section 1 142) offers the possibility of calculating the storage coefficient for different conditions with a shallow watertable This could be useful information for the drainage of areas prone to high capillary rise In general, p increases with increasing watertable depths The capillary reach in which equilibrium conditions exist is only active where the soil surface is nearby and when soil water is occasionally removed by evaporation For a depth greater than a certain critical value (which depends on the soil type), the drainable porosity can be approximated by the difference in 8 between field capacity and saturation The concept of drainable porosity is shown in Figure 1 112A In this figure, the soil-water content of a silty clay soil is shown by the line A-B for a watertable depth of 050 m, and by the line C-D for a watertable of 120 m The drainable porosity in this case is represented by the enclosed area ABCD (representing the change in soil-water content), divided by the change in watertable depth AD or p=- ABCD AD (1 122) Example 11 I Assume that the soil-water profile of Figure 1112 is in equilibrium (ie H = O) Then, according to Equation 1 119, h = -2, with z = O at the watertable and positive upward The pressure-head profile in thiscase is simply -z Pressure-head profiles for the two watertable depths are illustrated in Figure 1112B The soil-water content can now be determined graphically for each depth from the soil-water retention curve in Figure 1 112C The calculations are presented in Table 1 12 We divide the soil profile in discrete depth intervals of 010 m, and calculate the average difference in 8 between the first and the second watertable for each interval This average is multiplied by the interval depth, which yields the water content per interval, totalling 2805 mm We divide the total by 700 mm (ie the change in watertable depth), and find a drainable porosity p =

22 S depth in cm O O0 120 I 140 O O o water content O in cm3icm3 oresure head in cm volumetric soil water content O in cm3icm3 Figure 1112 A Soil-water-content profiles for equilibrium conditions with the watertable at 050 m, O,(z), and at 120 m, e,(z) The area enclosed by 8, (z), 8, (2) the soil surface, and AD represents the drainable porosity B Equilibrium pressure-head profiles for watertables at 050 and 120 m C Soil-water retention for a silty clay